Google Data Analytics Certificate Capstone Project
I. Background
Cyclistic is a fictitious bike-share company based in Chicago, I am undertaking a project that could significantly influence this company’s future. Cyclistic has been a success since its launch in 2016, growing to a fleet of 5,824 geotracked bicycles locked into a network of 692 stations across Chicago. The bikes can be unlocked from one station and returned to any other station in the system anytime.
with Bobby’s Bike Hike in Chicago, IL
Source: tripster.com
Chicago, once named the best bicycling city in the nation by Bicycling Magazine, has seen a surge in cycling popularity over the last decade. The city is well-equipped for cyclists, with almost every street featuring either a shared or buffered bike lane. Cyclistic’s bike-share program aligns perfectly with this culture, offering over 580 stations and 5,800 bikes across the city. The city hosts several cycling events, including the annual Bike the Drive event, which sees Lake Shore Drive closed to vehicular traffic, transforming it into a cyclist’s paradise.
II. Goals and Objectives
Cyclistic’s marketing strategy has so far focused on building general awareness and appealing to broad consumer segments. The company achieved this through the flexibile pricing plans: single-ride passes, full-day passes, and annual memberships. Customers who purchase single-ride or full-day passes are referred to as customers, while those who purchase annual memberships are Cyclistic subscribers.
Although the pricing flexibility helps the company attracts more customers, Moreno, Cyclistic’s marketing director, believes that maximizing the number of annual members will be key to the company’s future growth. Rather than creating a marketing campaign that targets all-new customers, Moreno sees a solid opportunity to convert casual riders into members.
III. Data & Data Processing
The data utilized for this case study was sourced from Cyclistic’s bike trip records for the first quarter of 2019. To facilitate a comprehensive spatial analysis, I supplemented the 2019 data with recent data from 2023, which includes geographic coordinates for each of Cyclistic’s bike stations in Chicago. The datasets employed for this study are both appropriate and unbiased, lending credibility to the analysis. These datasets were generously provided by Motivate International Inc. under this license. Importantly, these datasets respect user privacy, as they do not contain any confidential or personally identifiable information such as the names, contact details, or payment methods of the riders.
The initial steps in processing the data involved renaming columns for easier reference within the R programming environment. Following this, I created a new column to represent the day of the week when each trip commenced. Subsequently, I extracted the “time” data from the columns containing “datetime” information about the start and end of each customer trip. This allowed me to create a new column indicating the length (in minutes) of each trip taken.
Finally, for the sake of simplicity and to facilitate statistical analysis, I converted the data in the trip length column to minutes. This comprehensive and meticulous data processing approach ensures the accuracy and reliability of the subsequent analysis.
IV. Descriptive Statistics
Cyclistic 2019 dataset contains a total of 1,048,575 trips of which only 22.5% were taken by customers or casual as shown on figure1 on the left. The overall average ride length was 19.7 minutes, while customers and subscribers had an average of 41.5 and 13.4 minutes respectively. These statistics suggest that casual riders tend to take longer trips compared to their counterparts with annual membership with Cyclistic.
V. Data insights through visuals
The cycling habits of Cyclistic bike users exhibit significant variations based on their membership status. Users can be broadly categorized into two groups: casual customers who do not have a membership, and subscribers who hold an annual membership.
As depicted in the following figure, which represents the percentage of trips undertaken by customers and subscribers for each day of the week, a clear pattern emerges. Subscribers tend to take frequent trips throughout the day, every single day of the week, a behavior that is noticeably different from their non-member counterparts.
customer bike users take larger percentage of their trips
on weekends. .
Another intriguing observation from this graph is the distinct riding habits of customers and subscribers over the weekend and weekdays. Customers tend to ride more on Saturdays and Sundays, while subscribers are more active during the weekdays. This trend suggests that subscribers are more likely to use their bikes for commuting to work, while customers might be using the bikes more for leisure activities.
The graph on the right, labeled as Figure 3, shifts the focus to another aspect of user behavior: the length of the ride. The data suggests that customers tend to embark on longer rides, averaging up to 40 minutes each day of the week. In contrast, subscribers typically opt for shorter trips, averaging around 15 minutes. This difference in ride duration further underscores the divergent usage patterns between customers and subscribers, possibly reflecting their distinct needs and motivations for using the Cyclistic bike service.
The following graph to the left builds up on the idea that subscriber use Cyclistic bikes to commute to work. It illustrates the trend in the number of rides taken by subscribers at three-hour intervals throughout the day. Coincidentally the subscribers see their peak in the number of rides during the regular rush hours 6 to 8 am and 6 to 8 pm. This observation lends further evidence to the hypothesis that subscribers primarily use Cyclistic bikes for commuting to and from work.
during the regular rush hours 6 to 8 am and 6 to 8 pm.
Leveraging the available data for the starting and ending stations of each trip, I conducted an aggregate analysis to rank all the routes utilized by Cyclistic bike users. The results revealed the top five most frequented routes, which are as follows:
- 1st route: 1. Lake Shore Dr & Monroe St to Streeter Dr & Grand Ave, with a total of 2872 trips.
- 2 nd route: A round trip from Lake Shore Dr & Monroe St back to the same location, with 2720 trips.
- 3rd route: A round trip from Streeter Dr & Grand Ave back to the same location, with 2452 trips.
- 4th route: A round trip from Michigan Ave & Oak St back to the same location, with 1405 trips.
- 5th route: A round trip from Montrose Harbor back to the same location, with 1106 trips.
The subsequent graph provides a distribution of these trips among customers and subscribers. Interestingly, these routes are predominantly used by customer riders, with over 94 percent of customers using the first route.
This finding is particularly surprising and suggests that these routes may be popular for leisurely rides or tourist activities. In fact, Streeter Dr. & Grand Ave. is near a very popular pier in Chicago and a Museum, suggesting that people use the bikes for fun and to get to and between points of interest. Lake Shore Dr. & Monroe St. and Lake Shore Dr. & North Blvd. are also near Grant Park which is a great place to go bike riding. It also underscores the importance of understanding the different usage patterns of customers and subscribers, as this information can guide Cyclistic’s strategic planning, resource allocation, advatisement campains aimed at convering customers into subscribers. For instance, ensuring adequate bike availability on these popular routes could enhance the riding experience for customers and potentially convert them into annual subscribers.
is the most used route
Further dissecting the distribution of trip by riders’ age (Figure 2) and gender (Figure 3), a substantial portion of customer riders falls within the 21 to 30-year-old bracket. Additionally, an overwhelming majority (97%) of riders who did not disclose their gender are categorized as customer riders. This demographic revelation underscores the importance of customizing services and incentives tailored specifically to this age group. Aligning Cyclistic’s offerings more closely with the preferences and needs of this demographic holds immense value.
This data also suggests the necessity for Cyclistic to reconsider the approach to collecting gender-related information from riders. This could involve expanding the available options or reassessing the inclusion of this variable entirely. Rethinking how gender information is gathered will contribute to a more inclusive and reflective understanding of their diverse rider base, facilitating more targeted and comprehensive service enhancements.
My last insight from Cyclistic’s data is that most customer riders prefer electric bikes while subscribers use prefer classic ones as shown in figure 5. This insight suggest that Cyclistic should increase the number of bikes a the station that are frequently visited by customer riders.
VI. Concluding Thoughts
The analysis of Cyclistic’s bike trip data uncovers valuable insights pivotal for the company’s strategic direction. Distinguishing between casual riders and annual subscribers unveils distinct usage patterns. Subscribers predominantly engage in shorter rides, seemingly for daily commuting, evident from their peak ride times during rush hours. In contrast, casual riders opt for longer trips, especially on weekends, often visiting tourist attractions and leisure spots. Understanding these divergent behaviors is crucial for tailored marketing initiatives. Targeted conversion campaigns should spotlight the convenience and cost-effectiveness of annual memberships to entice casual riders into long-term subscriptions. Ensuring ample bike availability on popular routes frequented by casual riders, potentially for leisurely rides or tourist activities, remains essential. Customizing services and incentives for the influential 21 to 30-year-old demographic among casual riders could significantly boost membership conversion rates.
Moreover, the analysis underscores the necessity for inclusive data collection methods. Reevaluating the gathering of demographic information, particularly gender-related data, is imperative for reflecting the diverse rider base accurately. Cyclistic can enhance inclusivity and better align its services by revisiting how gender information is collected or expanding available options. Additionally, optimizing bike availability according to user preferences emerges as a key strategy. Considering the disparity in bike type preferences between casual riders (favoring electric bikes) and subscribers (preferring classic bikes), adjusting bike station offerings to cater to these preferences could significantly improve user experience and potentially drive membership conversions.
IX. Appendices
To see all Cyclistic bike stations on the map visit my Tableau Public account
Visit my Github for source code